Maritime characteristics make the progress of automatic operations in ships slow, especially compared to other means of transportation. This caused a great progressive deal of attention for Autonomy Degree (AD) of ships by research centers where the aims are to create a well-structured roadmap through the phased functional maturation approach to autonomous operation. Application of Maritime Autonomous Surface Ship (MASS) requires industries and authorities to think about the trustworthiness of autonomous operation regardless of crew availability on board the ship. Accordingly, this paper aims to prognose the health state of the conventional ships, assuming that it gets through higher ADs. To this end, a comprehensive and structured Hierarchal Bayesian Inference (HBI)-based reliability framework using a machine learning application is proposed. A machinery plant operated in a merchant ship is selected as a case study to indicate the advantages of the developed methodology. Correspondingly, the given main engine in this study can operate for 3, 17, and 47 weeks without human intervention if the ship approaches the autonomy degree of four, three, and two, respectively. Given the deterioration ratio defined in this study, the acceptable transitions from different ADs are specified. The aggregated framework of this study can aid the researchers in gaining online knowledge on safe operational time and Remaining Useful Lifetime (RUL) of the conventional ship while the system is being left unattended with different degrees of autonomy.